Coding or decoding method and communication device

Abstract

The invention provides a coding method, a decoding method and a communication device, in the coding method, a basis matrix of LDPC is designed by adopting a JSCC scheme, a state column is introduced into a verification part of the basis matrix, a first column of the verification part is a state column with high column weight, a second column is a column with a degree of 1, and the first column and the second column are different. According to the JSCC-LDPC decoding method and the JSCC-LDPC decoding system, the element '1' in the column with the degree of 1 and the element '2' in the state column are located in the same row of the basis matrix, and the combination of the column with the degree of 1 and the state column can improve the decoding threshold of the JSCC-LDPC and improve the decoding performance of the JSCC-LDPC.

Description

Technical Field

[0001] The embodiments of the present application relate to the field of information source and channel coding, and more specifically, to a coding or decoding method and a communication device. Background Art

[0002] In the field of channel coding, low-density parity check code (LDPC) is the most mature and widely used channel coding scheme. LDPC code has a performance close to the Shannon limit and has many advantages. Therefore, IEEE's 802.11n, 802.11ac, 802.11ax and other protocols propose to use LDPC as the standard channel coding scheme for wireless local area networks (WLAN). The 802.11ac / ax standard currently adopts 12 LDPC check matrices, of which there are 3 code lengths, and each code length supports 4 code rates. The encoding device selects the corresponding check matrix from the 12 check matrices for LDPC encoding based on the target code length and code rate.

[0003] However, the designs of existing cellular and wireless fidelity (WiFi) communication systems are all based on separate source channel coding (SSCC). The SSCC scheme cannot compress data at the physical layer. Once there is compressible data at the physical layer, it will lead to a waste of resources such as bandwidth and power. In addition, the decoding performance of the existing SSCC scheme still needs to be improved. Summary of the invention

[0004] The present application provides a coding or decoding method and a communication device, which can realize the function of compressing data at the physical layer and improve the decoding performance.

[0005] In a first aspect, a method for encoding or decoding is provided, which is applied to a communication device, such as a network device, a terminal device, a chip, a chip system, a hardware circuit, a software module, or a combination of a hardware circuit and a software module. The method may include:

[0006] Encoding or decoding based on a check matrix;

[0007] In which, the check matrix is ​​obtained by expanding the base matrix, the base matrix includes an information part and a check part, the check part includes a state column and a column with a degree of 1, the state column includes at least one element "2", the column with a degree of 1 includes an element "1", and an element "2" in the at least one element "2" and the element "1" are located in the same row of the base matrix.

[0008] It should be understood that when the method of the first aspect is applied to the encoding side, the communication device performs encoding based on the check matrix; if applied to the decoding side, the communication device performs decoding based on the check matrix.

[0009] In this application, the design of the base matrix adopts the joint source channel coding (JSCC) scheme, which can realize the function of compressing data at the physical layer. In addition, in order to achieve the compression of sparse sources, the JSCC-LDPC obtained by encoding based on the check matrix provided by this application only transmits the check bits on the channel, and all information bits are punctured before transmission. Introducing the state column in the check part of the base matrix can improve the decoding threshold and decoding performance of JSCC-LDPC.

[0010] By designing a basis matrix with a smaller dimension and then expanding the basis matrix to obtain a check matrix, the design complexity of the check matrix can be reduced.

[0011] In addition, in the state column of the base matrix, an element "2" in at least one element "2" and the element "1" are located in the same row of the base matrix, which can achieve a lower decoding threshold and greatly reduce the complexity of encoding and decoding.

[0012] On the encoding side, optionally, the method may further include: the communication device (the device corresponding to the encoding side) obtains a check matrix. Optionally. The method may further include: the communication device obtains an information sequence to be encoded. And, the communication device performs encoding based on the check matrix, including: the communication device encodes the information sequence to be encoded based on the check matrix to obtain coded bits. Optionally, the method may further include: the communication device outputs the coded bits.

[0013] On the decoding side, optionally, the method may further include: the communication device (device corresponding to the decoding side) obtains a check matrix. Optionally, the method may further include: the communication device obtains a sequence to be decoded. And, the communication device decodes based on the check matrix, including: the communication device decodes the sequence to be decoded based on the check matrix to obtain a decoded sequence. Optionally, the method may further include: the communication device outputs the decoded sequence.

[0014] In combination with the first aspect, in certain implementations of the first aspect, the state column of the base matrix also includes an element "1", and the number of elements "2" and elements "1" in the state column is close to that of the state column.

[0015] Such a design can achieve a compromise between the decoding threshold and the trap set to obtain an ideal decoding performance.

[0016] In combination with the first aspect, in certain implementations of the first aspect, the elements of the remaining columns of the base matrix except for the state column and the column with degree 1 include element "2", element "1" or element "0".

[0017] In combination with the first aspect, in some implementations of the first aspect, the elements of the last row of the remaining columns of the basis matrix except the state column and the column with degree 1 are 0.

[0018] This implementation method can reduce the design complexity of the check matrix to a large extent by reducing the design freedom.

[0019] In the above implementation, the elements of the state column of the base matrix are variable nodes corresponding to the punctured check bits.

[0020] In combination with the first aspect, in some implementations of the first aspect, expanding the base matrix to obtain the check matrix may include two-step expansion, wherein a first expansion factor corresponding to a first expansion of the two-step expansion is 2, and the first expansion includes:

[0021] Expanding the base matrix based on a first expansion factor to obtain a first matrix;

[0022] Among them, the element "2" in the base matrix is ​​expanded to a 2×2 all-1 matrix, the element "1" is expanded to a 2×2 unit matrix or an anti-diagonal square matrix, the element "0" is expanded to a 2×2 all-zero square matrix, and the first matrix includes the element "1" and the element "0".

[0023] In combination with the first aspect, in some implementations of the first aspect, a second expansion factor corresponding to the second expansion step of the two-step expansion is 32, and the second expansion step includes:

[0024] Expanding the first matrix based on the second expansion factor to obtain a check matrix;

[0025] The element "1" in the first matrix is ​​expanded into a 32×32 circulant permutation matrix P i , the element "0" in the first matrix is ​​expanded to a 32×32 all-zero square matrix, where 0≤i≤31 and i is an integer, P0 represents the unit matrix, P i It represents the square matrix obtained by cyclically shifting right each column vector in P0 by i positions.

[0026] In this implementation, the base matrix is ​​expanded in two steps through two expansion factors to obtain the check matrix. The design of the values ​​of these two expansion factors makes the dimensions of the base matrix in different source sparsity ranges 10 rows and 22 columns, 12 rows and 22 columns, and 14 rows and 22 columns, respectively, which can maximize the adaptation to the hardware in the WiFi standard and facilitate practical applications.

[0027] In combination with the first aspect, in some implementations of the first aspect, the dimension of the base matrix is ​​5 rows and 11 columns, the state column is the first column of the check part, the column with a degree of 1 is the second column of the check part, and the base matrix includes matrix 1, or a matrix obtained by row transformation or column transformation of matrix 1. Matrix 1 can refer to the embodiments of the specification and is not listed here.

[0028] Optionally, a basis matrix with a dimension of 5 rows and 11 columns can be applicable when the sparsity of the source belongs to the sparsity interval (0, 0.05]. In other words, when the sparsity of the source is in the sparsity interval (0, 0.05], the base matrix with a dimension of 5 rows and 11 columns given in this application can be used to encode by expanding the check matrix obtained.

[0029] In combination with the first aspect, in certain implementations of the first aspect, when the sparsity of the source belongs to the sparsity interval (0, 0.05], the check matrix is ​​such as matrix 2, or a matrix obtained by row transformation and / or column transformation of matrix 2, and matrix 2 refers to the embodiment of the specification.

[0030] In combination with the first aspect, in some implementations of the first aspect, the dimension of the base matrix is ​​6 rows and 11 columns, the state column is the first column of the check part, the column with a degree of 1 is the second column of the check part, and the base matrix includes matrix 3, or a matrix obtained by row transformation or column transformation of matrix 3. Matrix 3 can refer to the embodiments of the specification and is not listed here.

[0031] Optionally, a basis matrix with a dimension of 6 rows and 11 columns may be suitable for a source whose sparsity belongs to the sparsity interval (0.05, 0.11).

[0032] In combination with the first aspect, in some implementations of the first aspect, when the sparsity of the source belongs to the sparsity interval (0.05, 0.11], the check matrix is ​​such as matrix 4, or the matrix obtained by row transformation and / or column transformation of matrix 4, see the embodiment of the specification.

[0033] In combination with the first aspect, in some implementations of the first aspect, the dimension of the base matrix is ​​7 rows and 11 columns, the state column is the first column of the check part, the column with a degree of 1 is the second column of the check part, and the base matrix includes matrix 5, or a matrix obtained by row transformation or column transformation of matrix 5. Matrix 5 can refer to the embodiments of the specification and is not listed here.

[0034] Optionally, a basis matrix with a dimension of 7 rows and 11 columns may be suitable for a source whose sparsity belongs to the sparsity interval (0.11, 0.3).

[0035] In combination with the first aspect, in some implementations of the first aspect, when the sparsity of the source belongs to the sparsity interval (0.11, 0.3], the check matrix is ​​such as matrix 6, or the matrix obtained by row transformation and / or column transformation of matrix 6, and matrix 6 refers to the embodiment of the specification.

[0036] In the specific design of the base matrix, the specific design of the state column aims to achieve a compromise between the decoding threshold and the elimination of loops / traps. The combination of the column with degree 1 and the state column can achieve a lower decoding threshold and significantly reduce the encoding and decoding complexity.

[0037] In a second aspect, a communication device is provided, wherein the communication device has the function of implementing the method in the first aspect or any possible implementation thereof. The function can be implemented by hardware, or by hardware executing corresponding software implementation. The hardware or software includes one or more units corresponding to the above functions.

[0038] In a third aspect, the present application provides a communication device, comprising at least one processor, wherein the at least one processor is coupled to at least one memory, wherein the at least one memory is used to store computer programs or instructions, and the at least one processor is used to call and run the computer program or instructions from the at least one memory, so that the communication device executes the method in the first aspect or any possible implementation thereof.

[0039] In one example, the communication device described in the second aspect or the third aspect may be an encoding device or a decoding device.

[0040] In a fourth aspect, the present application provides a communication device, comprising a communication interface and a circuit, wherein the communication interface is used to receive an information sequence to be encoded (or called bits to be encoded) and input the information sequence to be encoded into the circuit; the circuit encodes the information sequence to be encoded based on the check matrix provided by the present application to obtain encoded bits; the communication interface is also used to output the encoded bits.

[0041] Exemplarily, the communication device of the fourth aspect is a coding device.

[0042] In a fifth aspect, a communication device is provided, comprising a communication interface and a circuit, wherein the communication interface is used to receive a sequence to be decoded and input the sequence to be decoded into the circuit; the circuit decodes the sequence to be decoded based on the check matrix provided in the present application to obtain a decoded sequence; and the communication interface is also used to output the decoded sequence.

[0043] Exemplarily, the communication device of the fifth aspect is a decoding device.

[0044] In a sixth aspect, the present application provides a computer-readable storage medium, in which computer instructions are stored. When the computer instructions are executed on a computer, the method in the first aspect or any possible implementation thereof is implemented.

[0045] In a seventh aspect, the present application provides a computer program product, comprising a computer program code, and when the computer program code is run on a computer, the method in the first aspect or any possible implementation thereof is implemented.

[0046] In an eighth aspect, the present application provides a wireless communication system, comprising a communication device as in any one of the second to fifth aspects, such as an encoding device and / or a decoding device. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 This is an example of the check matrix H of LDPC.

[0048] Figure 2 is an example of the Tanner graph corresponding to the check matrix H.

[0049] Figure 3 Schematic diagram of the shortening operation in the LDPC encoding process in WLAN.

[0050] Figure 4 Schematic diagram of the check matrix of the Double-LDPC scheme.

[0051] Figure 5 (a) and (b) are schematic diagrams of a communication system applicable to an embodiment of the present application.

[0052] Figure 6 A schematic flowchart of a method 600 for encoding or decoding provided by the present application.

[0053] Figure 7 to Figure 9 A schematic diagram of the structure of the base matrix provided in this application.

[0054] Fig.10 A comparison chart between the JSCC-EXIT graph analysis method provided in this application and the traditional JSCC-EXIT graph analysis method.

[0055] Fig.11 Schematic diagram of the basis matrix of NR LDPC.

[0056] Fig.12 Schematic diagram of the base matrix of NR LDPC and JSCC-LDPC provided in this application.

[0057] Fig.13 A schematic flow chart of designing a base matrix for the improved differential evolution method provided in this application.

[0058] Figure 14-16 This application provides a simulation comparison chart of the decoding performance of JSCC-LDPC and SSCC.

[0059] Fig.17 A schematic structural diagram of the communication device 1000 provided in this application.

[0060] Fig.18 A schematic structural diagram of the communication device 1100 provided in this application. DETAILED DESCRIPTION

[0061] The technical solutions in the embodiments of the present application will be described below in conjunction with the accompanying drawings.

[0062] In order to facilitate understanding of the technical solution of the present application, the relevant technologies involved in the embodiments of the present application are first introduced.

[0063] Data coding in existing cellular, wireless fidelity (WiFi) and other communication systems is based on the separate source channel coding (SSCC) theory. The encoder first performs source coding on the source data to be coded. Traditional source coding (for example, compression of sources such as images, videos, voices and text) does not have the ability to resist errors. Therefore, the physical layer needs to pass channel coding and combine retransmission and other mechanisms to enable the decoder to complete error-free recovery of the physical layer transmission bits. Source coding is generally completed at the application layer, and channel coding is generally completed at the physical layer. However, the SSCC-based scheme cannot compress data at the physical layer. Once there is compressible data at the physical layer, it will lead to waste of resources such as bandwidth and power, and reduce spectrum efficiency. In addition, the optimal performance of the SSCC-based scheme is based on Shannon's source channel separation theorem, which is based on infinite code length transmission without delay constraints. For finite code lengths, the academic community has proved that the transmission rate of joint source channel coding (JSCC) is better.

[0064] In order to make full use of the sparsity of original data such as voice, data, images and videos and improve transmission performance, JSCC is a potential solution that can simultaneously achieve source compression and channel protection. In addition, if there is source redundancy at the physical layer, JSCC can also further utilize its redundancy to achieve further compression and protection at the same time.

[0065] Low-density parity check code (LDPC) is the most mature and widely used channel coding scheme. The check matrix of LDPC is a sparse matrix, that is, the number of non-zero elements in the matrix is ​​much smaller than the number of zero elements, or the row weight and column weight of the matrix are very small compared to the code length. In 1981, Tanner used a graph to represent the codewords. This graphical representation is now called the Tanner graph, and it was proved that the sum-product algorithm can achieve the best decoding performance in the acyclic code graph. The Tanner graph corresponds to the check matrix one by one, and consists of two types of nodes. The first type is the variable node, which represents the codeword symbol; the second type of check node represents the check constraint relationship, and each node represents a check constraint relationship. Figure 1 An example of an LDPC check matrix H is given. Figure 2 The Tanner graph corresponding to the check matrix H is given, {V i} represents a variable node set, {C i} represents the check node set.

[0066] The LDPC used in the IEEE802.11ac / ax standard is quasi-cyslic LDPC (QC-LDPC). QC-LDPC is a widely used structured LDPC. Due to the unique structure of its check matrix, it can be implemented using a simple feedback shift register during encoding, which effectively solves the coding complexity problem of LDPC. At present, the LDPC channel decoder used in WiFi is fixed-length, and the dimensions of the base matrix used for different code rates are all [x, 24] in size, where x is the number of rows of the base matrix and x is a positive integer. The WiFi standard uses three code lengths and four code rates for check matrices, with a total of 12 check matrices. The check bit part of the 12 check matrices has the same structure. For example, the base matrix of the LDPC with code length N = 1944 and code rate R = 5 / 6 in 802.11ac is as follows:

[0067]

[0068] It can be seen that the dimension of the base matrix is ​​4×24. Each element in the base matrix represents a square matrix of order z=N / 24, where 0 represents the unit matrix, "-" represents the all-zero matrix, and the element P i represents the cyclic permutation matrix, i represents the cyclic shift value, 0≤i≤z-1, i is an integer. The following is a cyclic permutation matrix with a cyclic shift value of 1:

[0069]

[0070] The existing WLAN standards (such as IEEE802.11n / ac) use orthogonal frequency division multiplexing (OFDM) technology. The LDPC encoding module needs to encode the data bits and put them into an integer number of OFDM symbols, and these encoded bits must also fit into an integer number of LDPC codewords. To perform these steps, the system first calculates the minimum number of OFDM symbols N required for this transmission. SYM , and further according to N SYM Calculate the total number of coded bits N that can be stored in all OFDM symbols using the current coding and modulation scheme TCB ,N TCB =N CBPS ·N SYM , where N CBPS is the number of coded bits that can be stored in each OFDM symbol. Then, the system calculates the LDPC code length L used in the current transmission based on all the above calculation results. LDPC and the number of codewords N required CW For most combinations of the length of the data bits to be encoded and the coding modulation scheme, there are not enough data bits to fill the information bit part of the LDPC codeword, and a shortening operation needs to be performed before generating the check bits.

[0071] Figure 3 Figure 1 is a schematic diagram of the shortening operation in the LDPC coding process in WLAN. Shortening means that before the LDPC coding generates the check bits, a certain number of zeros are filled in the information bit part of the codeword, and these zeros are deleted after the coding generates the check bits. Figure 5 , step 1 is the data bits to be encoded; step 2 calculates the length L of the LDPC codeword LDPC and the number of codewords N CW ; Step 3 shortens the data bits; Step 4 generates check bits using the information bits and the 0s to be shortened, and finally deletes the 0s to be shortened.

[0072] The LDPC channel coding under the current WiFi standard is designed for compressed data, so there is no compression function. Once there is compressible data in the physical layer, it will cause a waste of resources such as bandwidth and power.

[0073] The Double-LDPC scheme is a currently known JSCC scheme based on LDPC. This scheme cascades two LDPCs. The source LDPC is responsible for source compression, and the channel LDPC is responsible for channel protection. Decoding can be done in sequence or in parallel.

[0074] Figure 4is a schematic diagram of the check matrix of the Double-LDPC scheme. Figure 4 It can be seen that the source LDPC part of the check matrix (i.e. H SC ) and the LDPC part of the channel (i.e. H CC ) are relatively independent in design, and the division of labor between the two is clear. Even if an improvement plan is proposed later, the connection relationship between the two matrices is relatively fixed, and the code rate allocation of the source coding part and the channel coding part has a greater impact on the performance and there is a compromise. This solution has a large change to the WiFi standard hardware, which is not easy to implement, and the performance needs to be improved.

[0075] Based on the above technical status, this application provides a coding method, which aims to provide a JSCC-LDPC check matrix, which can realize the compression function of the physical layer and can maximize the adaptation of existing WiFi hardware. In addition, compared with some existing coding schemes, the decoding performance is also improved.

[0076] The technical solution provided by this application is introduced below.

[0077] The technical solution of the present application can be applied to various wireless communication systems, including but not limited to: the 5th generation (5G) system or new radio (NR) system, long term evolution (LTE) system, long term evolution-advanced (LTE-A) system, LTE frequency division duplex (FDD) system, LTE time division duplex (TDD) system, etc., and can also be applied to future communication systems, such as the sixth generation mobile communication system. In addition, it can also be applied to device to device (D2D) communication, vehicle-to-everything (V2X) communication, machine to machine (M2M) communication, machine type communication (MTC), and Internet of Things (IoT) communication system or other communication systems, etc. In addition, the technical solution of the present application can be applied to WLAN scenarios, for example, it can be applied to the 802.11 system standards of the Institute of Electrical and Electronics Engineers (IEEE), such as the 802.11a / b / g standard, 802.11n standard, 802.11ac standard, 802.11ax standard, 802.11be standard or its next generation standard, without limitation. The above communication system applicable to the present application is only an example, and the communication system applicable to the present application is not limited thereto.

[0078] See also Figure 5 , Figure 5 (a) and (b) are schematic diagrams of a communication system applicable to an embodiment of the present application. The communication system includes an encoding end and a decoding end. The encoding end is not limited to one or more, and the decoding end is not limited to one or more. Exemplarily, one of the encoding end and the decoding end can be an access point (AP) in a WLAN scenario, and the other can be a station (STA). The embodiment of the present application can be applicable to wireless communication between AP and STA. Optionally, it can include communication scenarios such as AP communicating with a single STA and AP communicating with multiple STAs. Communication between AP and multiple STAs can also include downlink transmission of signals sent by AP to one or more STAs, and uplink transmission of signals sent by one or more STAs to AP. Figure 5(a) shows a scenario where an AP communicates with a single STA. Figure 5 (b) shows a scenario where an AP communicates with multiple STAs. The encoding end in the following embodiments may be Figure 5 The STA in uplink transmission shown in may also be an AP in downlink transmission. In addition, the technical solution of the present application may also be applied to similar non-standard wireless transmission systems without limitation.

[0079] The encoding device and decoding device in the embodiment of the present application can be a terminal device, including various communication kits with wireless communication functions (the communication kits can include, for example, antennas, power supply templates, cables, and WiFi modules, etc.), handheld devices, vehicle-mounted devices, or other processing devices connected to a wireless modem, and can specifically refer to user equipment (UE), users, access terminals, user units, user stations, mobile stations, mobile stations, remote stations, remote terminals, mobile devices, user terminals, terminals, wireless communication devices, user agents, user devices, wireless modems, machine type communication devices, or other processing devices connected to wireless modems. It can also be a virtual reality (VR) terminal device, an augmented reality (AR) terminal device, a wireless terminal in industrial control, a terminal in self-driving, a terminal in remote medical, a terminal in smart grid, a terminal in transportation safety, a terminal in smart city, a terminal in smart home, or a terminal device in a future communication network. Of course, the terminal device in this application may also refer to a chip, a modem, a system on a chip (SoC) that is mainly responsible for related communication functions in the device, or a communication platform that may include a radio frequency (RF) part, etc.

[0080] The encoding device and decoding device in the embodiment of the present application may also be a network device, which may include but is not limited to: a next-generation base station (gNodeB, gNB) in a fifth-generation (5G) communication system, a base station in a sixth-generation (6G) mobile communication system, a base station in a future mobile communication system, an access point (AP) in a wireless fidelity (WiFi) system, a wireless relay node, a wireless backhaul node, a transmission point (TP) or a transmission reception point (TRP), an evolved node B (evolved node B, eNB) in a long-term evolution (LTE) system, a network device in a non-terrestrial network (NTN) communication system, etc. The network device may also be one or a group (i.e., multiple) antenna panels of a base station. In addition, the network device may also be a network node constituting a gNB or TP, such as a baseband unit (BBU), a centralized unit (CU), a distributed unit (DU), or a radio unit (RU). Alternatively, the network device may also be a device-to-device (D2D) communication system, a machine-to-machine (M2M) communication system, an Internet of Things (IoT), an Internet of Vehicles communication system, or a device that performs network-side functions in other communication systems, without limitation.

[0081] In the embodiments of the present application, the device for realizing the terminal function may be a terminal, or a device capable of supporting the terminal to realize the function, such as a chip system or a chip, which may be installed in the terminal. In addition, the device for realizing the function of a network device may be a network device, or a device capable of supporting the network device to realize the function, such as a chip system or a chip, which may be installed in the network device. Optionally, the chip system may be composed of a chip, or may include a chip and other discrete devices.

[0082] Figure 6A schematic flowchart of a method 600 for encoding / decoding in a wireless system provided in the present application, wherein: step 610 can be performed by an encoding device, or by a device installed in the encoding device (such as a chip, a chip system or a circuit, etc.); step 650 can be performed by a decoding device, or by a device installed in a decoding device (such as a chip, a chip system or a circuit, etc.).

[0083] The following uses an encoding device as an example to illustrate the encoding side method.

[0084] 610. The encoding device performs encoding based on a check matrix.

[0085] The check matrix provided in this application is obtained by expanding the base matrix.

[0086] Specifically, the base matrix includes an information part and a check part. The check part includes a state column and a column with a degree of 1, the state column includes at least one element "2", the column with a degree of 1 includes an element "1", and an element "2" of the at least one element "2" and the element "1" are located in the same row of the base matrix, or in other words, the element "1" in the column with a degree of 1 is connected to an element "2" in the state column.

[0087] Furthermore, the state column also includes element "1", and the number of elements "2" and "1" in the state column is close. The element "1" in the column with degree 1 is connected to the element "2" in the state column, but not to the element "1" or located in the same row of the basis matrix.

[0088] It should be noted that the elements of the state column in this application are variable nodes corresponding to the check bits that will be punctured during transmission. During the research process of this application solution, it was found that introducing the state column in the check part of the base matrix can further improve the decoding threshold.

[0089] Optionally, the elements of the remaining columns in the base matrix except the state column and the column with degree 1 include the element "2", the element "1" or the element "0". That is, the value of the element at any position in the remaining columns in the base matrix except the state column and the column with degree 1 can be designed to be 2, 1 or 0.

[0090] Optionally, as an implementation manner, the last row of the remaining columns except the state column and the column with degree 1 in the base matrix is ​​an element "0".

[0091] Alternatively, in various embodiments of the present application, the element "x" may also mean that the value of the element is "x". For example, the column with a degree of 1 includes element 1, which means that the column with a degree of 1 includes an element whose value is 1. The state column includes at least one element "2", which means that the state column includes at least one element, and the value of the at least one element is 2.

[0092] Optionally, expanding the base matrix to obtain the check matrix may include two-step expansion.

[0093] Specifically, the first step of the two-step expansion corresponds to the first expansion factor, and the second step corresponds to the second expansion factor. The expansion process can be: expanding the base matrix based on the first expansion factor to obtain the first matrix; expanding the first matrix based on the second expansion factor to obtain the check matrix. Based on the above description, it can be seen that, in general, the base matrix includes three elements, namely element "2", element "1" and element "0". If the first expansion factor is 2, then in the first step of the expansion, the element "2" in the base matrix is ​​expanded to a 2×2 all-1 matrix, the element "1" is expanded to a 2×2 unit matrix or an anti-diagonal square matrix, and the element "0" is expanded to a 2×2 all-zero square matrix.

[0094] After the first step of expansion, the base matrix is ​​expanded to the first matrix, which includes two elements, element "1" and element "0". The second expansion factor is 32, so in the second step of expansion, the element "1" in the first matrix is ​​expanded to a 32×32 circulant permutation matrix P i , the element "0" in the first matrix is ​​expanded to a 32×32 square matrix of all zeros, where 0≤i≤31 and i is an integer, P0 represents the unit matrix, P i It represents the square matrix obtained by cyclically shifting right each column vector in P0 by i positions.

[0095] Optionally, for different information source sparsities, the present application provides several different dimensions of the basis matrix.

[0096] (1) The dimension of the basis matrix is ​​5 rows and 11 columns.

[0097] One of the columns of the check part of the base matrix is ​​a state column, one of the columns of the check part is a column with a degree of 1, the state column includes at least one element "2", the column with a degree of 1 includes an element "1", and an element "2" of the at least one element "2" and the element "1" are located in the same row of the base matrix. Optionally, the elements of the remaining columns of the base matrix except the state column and the column with a degree of 1 are elements "2", elements "1" or elements "0". As an example, the last row of the remaining columns of the base matrix except the state column and the column with a degree of 1 is an element "0".

[0098] As mentioned above, the number of elements "2" and "1" in the status column is similar. For example, the status column is [1 1 1 2 2] T , it can be seen that the state column includes 5 elements in total, the number of elements "2" and "1" accounts for 2 and 3 respectively, which is close to half. In addition, the column with degree 1 is [0 0 0 0 1]T It can be seen that the element "1" in the degree 1 column and an element "2" in the state column are located in the same row in the basis matrix.

[0099] Based on the above design ideas, in one case, the dimension of the base matrix is ​​5 rows and 11 columns. For example, the state column is the first column of the check part, and the column with degree 1 is the second column of the check part. The design of the base matrix can be as follows: Figure 7 shown.

[0100] As an example, the present application provides a base matrix such as Matrix 1, or a matrix obtained by performing row transformation and column transformation on Matrix 1.

[0101] Matrix 1

[0102]

[0103] Based on matrix 1 and the above two-step expansion process, the check matrix obtained by expanding the base matrix may be specifically matrix 2. In addition, the check matrix may also be a matrix obtained by performing row transformation and / or column transformation on matrix 2.

[0104] Matrix 2

[0105]

[0106] The element "-1" in the matrix 2 represents a z×z all-zero matrix, the element "0" represents a z×z unit square matrix, and the element P i represents a cyclic permutation matrix of z×z, i represents a cyclic shift value, 0≤i≤z-1, i is an integer, and z represents the second expansion factor. As an example, the second expansion factor z=32. In addition, the element "-1" in the check matrix can also be replaced by other symbols, for example, the symbol "-" is used to represent an all-zero matrix. This description is also applicable to other check matrices, and will not be repeated below.

[0107] Optionally, the above matrix 1 or matrix 2 can be used for encoding with a target code rate R=1.5.

[0108] In addition, due to the large dimension of the check matrix, a cyclic shift permutation matrix is ​​generally used to represent the check matrix, and the two are equivalent. For example, each element in matrix 2 corresponds to a cyclic permutation matrix or an all-zero matrix or a unit matrix with a dimension of 32×32. After matrix 2 is expanded, the corresponding matrix is ​​the check matrix. Therefore, the dimension of the check matrix can be the dimension of the matrix after matrix 2 is expanded. But it can also be said that matrix 2 is a check matrix. Therefore, it can also be said that the dimension of matrix 2 is also the dimension of the check matrix. These replacement statements known to those skilled in the art should not impose any limitations on the check matrix in the embodiments of the present application. This will not be repeated in the following other embodiments.

[0109] (2) The dimension of the basis matrix is ​​6 rows and 11 columns.

[0110] One of the columns of the check part of the base matrix is ​​a state column, one of the columns of the check part is a column with a degree of 1, the state column includes at least one element "2", the column with a degree of 1 includes an element "1", and an element "2" of the at least one element "2" and the element "1" are located in the same row of the base matrix. Optionally, the elements of the remaining columns of the base matrix except the state column and the column with a degree of 1 are elements "2", elements "1" or elements "0". As an example, the last row of the remaining columns of the base matrix except the state column and the column with a degree of 1 is an element "0".

[0111] For example, the status column is [1 2 1 2 1 2] T , the state column includes 6 elements, the number of elements "2" and "1" are 3 and 3 respectively, each accounting for half. In addition, the column with degree 1 is [0 0 0 0 0 1] T . An element "1" in the degree 1 column and an element "2" in the state column are located in the same row in the basis matrix.

[0112] Based on the above design ideas, in one case, the dimension of the base matrix is ​​6 rows and 11 columns. For example, the state column is the first column of the check part, and the column with degree 1 is the second column of the check part. The design of the base matrix can be as follows: Figure 8 shown.

[0113] As an example, the present application provides a base matrix such as Matrix 3, or a matrix obtained by performing row transformation and / or column transformation on Matrix 3.

[0114] Matrix 3

[0115]

[0116] Based on matrix 3 and the above two-step expansion process, the check matrix obtained by expanding the base matrix can be specifically matrix 4, or a matrix obtained by performing row transformation and / or column transformation on matrix 4, without limitation.

[0117] Matrix 4

[0118]

[0119] The element "-1" in the matrix 4 represents a z×z all-zero matrix, the element "0" represents a z×z unit square matrix, and the element P i represents a z×z cyclic permutation matrix, i represents a cyclic shift value, 0≤i≤z-1, i is an integer, and z represents the second extension factor. As an example, the second extension factor z=32.

[0120] Optionally, the above matrix 3 or matrix 4 can be used for encoding with a target code rate R=1.

[0121] (3) The dimension of the basis matrix is ​​7 rows and 11 columns.

[0122] One of the columns of the check part is a state column, one of the columns of the check part is a column with a degree of 1, the state column includes at least one element "2", the column with a degree of 1 includes an element "1", and an element "2" of the at least one element "2" and the element "1" are located in the same row of the base matrix. Optionally, the element at any position of the remaining columns of the base matrix except the state column and the column with a degree of 1 is an element "2", an element "1" or an element "0". As an example, the last row of the remaining columns of the base matrix except the state column and the column with a degree of 1 is an element "0".

[0123] For example, the status column is [1 1 1 1 2 2 2] T , the state column includes 7 elements in total, the number of elements "1" and elements "2" account for 4 and 3 respectively, which is close to half. In addition, the column with degree 1 is [0 0 0 0 0 01] T . An element "1" in the degree 1 column and an element "2" in the state column are located in the same row in the basis matrix.

[0124] Based on the above design ideas, in one possible case, the dimension of the base matrix is ​​7 rows and 11 columns. For example, the state column is the first column of the check part, and the column with degree 1 is the second column of the check part. The design of the base matrix can be as follows: Fig. 9 shown.

[0125] As an example, the present application provides a base matrix such as matrix 5, or a matrix obtained by performing row transformation and / or column transformation on matrix 5.

[0126] Matrix 5

[0127]

[0128] Based on matrix 5 and the above two-step expansion process, the check matrix obtained by expanding matrix 5 can be specifically matrix 6, or a matrix obtained by performing row transformation and / or column transformation on matrix 6, without limitation.

[0129] Matrix 6

[0130]

[0131] The element "-1" in the matrix 6 represents a z×z all-zero matrix, the element "0" represents a z×z unit square matrix, and the element P irepresents a z×z cyclic permutation matrix, i represents a cyclic shift value, 0≤i≤z-1, i is an integer, and z represents the second extension factor. As an example, the second extension factor z=32.

[0132] Optionally, the above matrix 5 or matrix 6 can be used for encoding with a target code rate R=2 / 3.

[0133] The above-mentioned base matrices of different dimensions can be applied to encoding under different source sparsities. As an example, when the source sparsity belongs to the sparsity interval (0, 0.05], matrix 1 can be used as the base matrix, or matrix 2 can be used as the check matrix for encoding. In this case, the target coding rate can be R = 1.5. When the source sparsity belongs to the sparsity interval (0.05, 0.11], matrix 3 can be used as the base matrix, or matrix 4 can be used as the check matrix for encoding. In this case, the target coding rate can be R = 1. When the source sparsity belongs to the sparsity interval (0.11, 0.3], matrix 5 can be used as the base matrix, or matrix 6 can be used as the check matrix for encoding. In this case, the target coding rate can be R = 2 / 3.

[0134] It should be understood that since the check matrix is ​​usually obtained by expanding the base matrix, the above embodiment gives an example of the base matrix, and the base matrix is ​​expanded in two steps to obtain the check matrix of JSCC-LDPC. Optionally, the check matrix can also be designed directly. In this case, the embodiments of the present application provide check matrices such as the above-mentioned matrix 2, matrix 4 or matrix 6, or matrices obtained by row transformation and / or column transformation of these matrices, without limitation.

[0135] Optionally, the encoding method may further include steps 620 to 630.

[0136] 620. The encoding device obtains an information sequence to be encoded (or referred to as information bits).

[0137] The encoding device performs encoding based on the check matrix, including the encoding device encoding the information sequence to be encoded based on the check matrix to obtain the encoded bits.

[0138] 630. The encoding device outputs the encoded bits.

[0139] Optionally, the encoding method may further include step 640 .

[0140] 640. The encoding device obtains a check matrix.

[0141] There are many ways for the coding device to obtain the check matrix. Exemplarily, the coding device reads the check matrix from the memory; or the processing module (such as a processor, circuit, chip or chip system, etc.) of the coding device reads the base matrix from the memory, and determines (or generates) the check matrix based on the base matrix. Optionally, the memory can be integrated with the processing module or exist physically separately, without limitation.

[0142] The above is the method on the encoding side.

[0143] The decoding side method is described below using a decoding device as an example.

[0144] 650. The decoding device performs decoding based on the check matrix.

[0145] It should be understood that the check matrix used by the decoding device for decoding in step 650 is the same as the check matrix used by the encoding device for encoding.

[0146] Optionally, the decoding method may further include steps 660 to 670 .

[0147] 660. The decoding device obtains a sequence to be decoded.

[0148] For example, the decoding device receives the sequence to be decoded from the encoding device via an air interface.

[0149] Exemplarily, the decoding device performs decoding based on a check matrix, including: the decoding device decodes the sequence to be decoded based on the check matrix to obtain a decoded sequence.

[0150] 670. The decoding device outputs the decoded sequence.

[0151] Optionally, the decoding method may further include step 680 .

[0152] 680. The decoding device obtains a check matrix.

[0153] It should be understood that step 680 may be located before step 650. There are many possible ways for the decoding device to obtain the check matrix. Exemplarily, the decoding device receives the check matrix encoded by the encoding device. Alternatively, the decoding device reads the check matrix for this decoding from a plurality of pre-stored check matrices based on the indication (e.g., index) of the encoding device. Alternatively, the decoding device reads the base matrix from the memory, and generates the check matrix based on the base matrix, etc., without limitation.

[0154] It should be noted that Figure 6 The serial numbers of the steps are only used to distinguish different steps so as to clearly describe the technical solution. The size of the serial numbers of the steps does not mean the order of execution. The execution order of the steps should be determined by the internal logic of the encoding method / decoding method.

[0155] The above describes the base matrix or check matrix of the JSCC-LDPC scheme given in this application.

[0156] It can be understood that the design of the base matrix or the check matrix takes into account many factors. The above embodiments provide the check matrix and base matrix of the JSCC-LDPC provided by this application. The following is a method for designing the check matrix and base matrix of the JSCC-LDPC. The check matrix obtained by the design method provided by this application can improve the decoding performance under the premise of realizing the compression function of the physical layer.

[0157] In summary, the specific design process of the JSCC-LDPC check matrix for WLAN in this application mainly includes two steps:

[0158] The first step is the design of the base model diagram.

[0159] In the design of the base model graph, the present application first designs a calculation method for the JSCC decoding threshold (hereinafter referred to as the decoding threshold). It is known that the decoding threshold is a theoretical standard for judging the quality of the JSCC base model graph. By analyzing the calculation method of the decoding threshold, a calculation method that can obtain an accurate decoding threshold is adopted. Furthermore, based on the basic structure and constraints that the base model graph provided by the present application should meet, and combined with the judgment criteria of the decoding threshold, a specific base model graph is designed for the source sparsity p and the target bit rate.

[0160] The second step is the expansion of the base model diagram.

[0161] The expansion of the base model graph refers to the expansion of the designed base matrix into the final check matrix using a circulant permutation matrix (CPM) and an all-zero square matrix. It should be pointed out that the error floor of traditional JSCC coding is difficult to reduce. The error floor of traditional channel coding is mainly dominated by the trap set. This application takes into account the difference between the impact of the trap set of JSCC-LDPC on the decoding performance and the impact of traditional channel coding on the decoding performance. Therefore, in the second step of the base model graph expansion, the trap set of the code in the JSCC scenario is analyzed to evaluate the trap set performance of the candidate check matrix obtained by the expansion, so as to determine the final check matrix of JSCC-LDPC.

[0162] The above two steps can be roughly summarized as the following design process:

[0163] Step 1: Design of the base model graph (or base matrix), including the following aspects:

[0164] 1.1. The calculation method of the decoding threshold is used to judge the quality of the base mode graph or degree distribution;

[0165] 1.2. The basic structure and constraints that the base model diagram needs to comply with;

[0166] 1.3. Construction of a base model graph: Based on the decoding threshold (calculated using the improved method for calculating the decoding threshold provided in the present application), the basic structure of the base model graph and the constraints, a specific base model graph is constructed.

[0167] Step 2: The expansion of the base model diagram mainly involves the following aspects:

[0168] 2.1. Trap set analysis: the criterion for judging decoding performance, especially the error flat area;

[0169] 2.2. Extension factor design: Expand the designed base mode graph into the check matrix of JSCC-LDPC;

[0170] 2.3. Trap set evaluation: Quickly locate the dominant trap set to determine the final check matrix.

[0171] The above aspects are introduced in detail below.

[0172] 1.1. Calculation method of decoding threshold.

[0173] The present application adopts an improved JSCC-extrinsic information transfer chart (EXIT) chart analysis method for calculating the decoding threshold. It should be understood that the improved JSCC-EXIT is different from the traditional JSCC-EXIT chart analysis method.

[0174] Fig.10 A comparison chart of the improved JSCC-EXIT graph analysis method provided in this application and the traditional JSCC-EXIT graph analysis method. Fig.10 , the traditional JSCC-EXIT analysis approximates the biased signal source and the transmission channel as a single white Gaussian noise (AWGN) channel, and in the research process of this application, it is found that this will lead to inaccurate calculation of the mutual information of the JSCC-EXIT graph. Therefore, this application proposes an improved JSCC-EXIT graph analysis method, which processes the BSC channel and the transmission channel that are equivalent to the biased signal source part separately, so that the mutual information of the iterative decoding process can be tracked more accurately, and thus the decoding threshold can be obtained more accurately. Among them, the decoding threshold of the base mode graph refers to the minimum signal-to-noise ratio (SNR) or Eb / N0 required for the LDPC code set corresponding to the base mode graph to be successfully decoded when the code length is infinite, that is, the smaller the decoding threshold, the closer the LDPC code set corresponding to the base mode graph is to the channel capacity when the code length tends to be infinite, and the better the performance.

[0175] 1.2. The basic structure and constraints of the base model diagram.

[0176] In traditional JSCC-LDPC coding, for example, for NR LDPC suitable for channel coding, except for the coded bits corresponding to the state nodes, the remaining coded bits are transmitted through the channel, and the optimization of the base model graph (or base matrix) is almost the optimization of the information part of the base matrix, and the structure of the check part of the base matrix is ​​relatively fixed. Among them, the state node mentioned in NRLDPC refers to the variable node corresponding to the information bit punctured in the channel coding scenario.

[0177] Fig.11 Schematic diagram of the basis matrix of NR LDPC.

[0178] like Fig.11 ,Since the structure of the check part of the base model graph (or base matrix) is fixed, the optimization of the base model graph in the case of channel coding is mainly to optimize the information part in the base matrix. For channel coding, the check part in the fixed base matrix does not affect the overall performance, so all coded bits (except the bits corresponding to the state nodes) pass through the same channel.

[0179] In order to achieve compression of sparse information sources, the JSCC-LDPC proposed in this application only transmits check bits on the channel, and all information bits are punctured before transmission, so both the source and channel parts should be optimized.

[0180] In summary, the present application proposes a basic structure and four constraints for the base matrix of JSCC-LDPC. For the convenience of description, the basic structure and the four conditions are respectively denoted as S1 and C1~C4.

[0181] S1: Introducing state nodes in the check part of the base matrix can improve the decoding threshold of JSCC and the actual performance of JSCC-LDPC.

[0182] It should be noted that the base matrices of JSCC-LDPC and NR LDPC provided in this application are different. Fig.12 Schematic diagram of the base matrix of NR LDPC and JSCC-LDPC provided in this application. Fig.12 ,In the basis matrix of NR LDPC, the state nodes are located in the information part, while in the basis matrix of JSCC-LDPC, the state nodes are located in the check part.

[0183] C1: For the base matrix constructed in this application, the square matrix corresponding to the check part after expansion is full rank to ensure that the source bits are encoded without bit rate loss. Specifically, the even numbers in the base matrix can be set to zero and the odd numbers can be set to 1, thereby ensuring that the obtained binary square matrix is ​​full rank under binary operations. It should be noted that the even numbers in the base matrix specifically refer to the element "0" and the element "2", and the odd numbers in the base matrix specifically refer to the element "1".

[0184] C2: Control the maximum row weight and maximum column weight in the basis matrix to control the density of 1 in the basis matrix to ensure the performance of the basis matrix under medium and short code lengths.

[0185] C3: Control the number of variable nodes with degree 2 to ensure that the minimum distance of the constructed basis matrix can grow linearly with the code length, thereby better controlling the performance of the error floor.

[0186] C4: It is necessary to store multiple base matrices with better decoding thresholds for further optimization, so as to determine the final base matrix from the multiple base matrices with better decoding thresholds. The multiple base matrices with better decoding thresholds can also become candidate base matrices.

[0187] The reason for designing this constraint is that for short codes, the decoding threshold cannot completely determine the final performance of the actual code, because the complex subgraph structure in the factor graph of the short code will seriously affect the decoding performance. Therefore, storing multiple basis matrices with better thresholds can better select basis matrices with excellent performance under medium and short code lengths.

[0188] Based on the above introduction, the method for constructing a base matrix provided in this application can be summarized as the following process: based on the basic structure and constraints of the base matrix provided in this application, multiple candidate base matrices are obtained; based on the multiple candidate base matrices, the base matrix is ​​determined (i.e. corresponding to the above-mentioned constraint C4).

[0189] Among them, the basic structure of the basis matrix (i.e. S1) includes:

[0190] The base matrix includes an information part and a check part, the check part includes a state column, and the state nodes included in the state column are variable nodes corresponding to the check bits punctured during transmission;

[0191] The constraints on the basis matrix include:

[0192] After expansion, the check part of the basis matrix satisfies the full rank (i.e., C1) under the modulo-2 operation;

[0193] The maximum row weight and the maximum column weight of the base matrix meet the first setting condition, which is set based on the performance of the waterfall region and the error flat region of JSCC-LDPC (i.e., C2);

[0194] The number of variable nodes of degree 2 of the base matrix meets the second setting condition, which is set based on the performance of the error floor region of JSCC-LDPC (ie, C3).

[0195] Based on the above introduction, compared with the traditional channel coding LDPC, the JSCC-LDPC for WLAN in this application needs to meet more structural characteristics and constraints, so the check part and the information part of the base matrix have different designs from the traditional base matrix. When designing a specific base matrix, the temporary or final base matrix generated by each design step needs to meet the above constraints C1 to C3 on the basis of the basic structure described in S1 above, and the generation of the base matrix meets the constraint C4.

[0196] 1.3. Construction of basis matrix.

[0197] Specifically, a method based on differential evolution can be used to iteratively obtain multiple candidate basis matrices, wherein the differential evolution method includes using the above-mentioned improved JSCC-EXIT graph analysis method to calculate the decoding threshold, and the temporarily generated basis matrix or candidate basis matrix must satisfy the basic structure S1 and constraints C1~C3.

[0198] Fig.13 The schematic flow chart of designing the basis matrix of the improved differential evolution method provided in this application is as follows. Fig.13 , determine the dimension of the basis matrix according to the sparsity p of the information bits and the target code rate. Based on the determined dimension of the basis matrix, randomly generate M basis matrices, and the M basis matrices all satisfy the basic structure S1. Determine whether the M basis matrices all satisfy the constraints C1~C3. If not, regenerate M basis matrices until each of the generated M basis matrices satisfies the basic structure S1 and the constraints C1~C3. Afterwards, use the improved differential evolution method to iterate the M basis matrices, and use the improved JSCC-EXIT graph analysis method to calculate the decoding threshold during the iteration process. Further determine whether all the basis matrices generated by the iteration satisfy the constraints C1~C3. It should be understood that although the randomly generated M basis matrices satisfy the constraints C1 to C3, the basis matrices obtained through the iteration of differential evolution do not necessarily satisfy the constraints C1 to C3. Therefore, it is necessary to ensure that each basis matrix generated by the iteration still satisfies the constraints C1 to C3, otherwise continue to iterate through the improved differential evolution method until each basis matrix obtained through the iteration of the improved differential evolution method satisfies the constraints C1 to C3. If the improved differential evolution algorithm has not converged or has not reached the maximum number of iterations, continue to iterate; when the improved differential evolution algorithm converges or reaches the maximum number of iterations, output the X basis matrices generated by the iteration, which are referred to as X candidate basis matrices in this article, and the basis matrix of JSCC-LDPC can be determined from the X candidate basis matrices.

[0199] Before designing the specific basis matrix of JSCC-LDPC, the present application divides the source sparsity into several typical intervals, namely (0, 0.05], (0.05, 0.11], (0.11, 0.3] and (0.3, 0.5]. These four sparsity intervals can cover all different source sparsities. Therefore, the present application only needs to design the basis matrix with source sparsities of 0.05, 0.11 and 0.30 respectively. The basis matrix corresponding to different source sparsities can be expanded to obtain the corresponding check matrix. In practical applications, the corresponding check matrix is ​​selected according to the current actual source sparsity p1. The corresponding relationship can be as follows:

[0200] When p1∈(0, 0.05], select the check matrix corresponding to p1=0.05;

[0201] When p1∈(0.05, 0.11], select the check matrix corresponding to p1=0.11;

[0202] When p1∈(0.11, 0.3], select the check matrix corresponding to p1=0.30;

[0203] When p1∈(0.3, 0.5], the traditional check matrix for equal probability sources is selected.

[0204] In order to adapt to the hardware in the existing WiFi standard, while considering the source sparsity and target bit rate design, this application designs the basis matrix with source sparsity of 0.05, 0.11, and 0.30 respectively. The ideas are as follows:

[0205] The base matrix includes an information part and a check part. The check part includes a state column, which is also called a puncturing column. In addition, during the research process of the present application scheme, it was found that the combination of the state node and the node with degree 1 can better adapt to the hybrid automatic repeat request (HARQ). Therefore, the check part also includes a column with degree 1. The state column includes at least one element "2", the column with degree 1 includes the element "1", and the element "1" in the column with degree 1 and one of the elements "2" in the at least one element "2" in the state column are located in the same row in the base matrix. In addition, the values ​​of the last elements of the remaining columns in the base matrix except the state column and the column with degree 1 can all be 0. Exemplarily, for different source sparsities, the base matrix structure of JSCC-LDPC provided in the present application is as follows: Figure 7-Figure 9 as shown in .

[0206] It should be noted that Figures 7 to 9The base matrix structure shown in is only an example. Optionally, the state column can be any column of the check part, and the column with degree 1 can be any column of the check part. Figures 7 to 9 In the example, the state column is used as the first column of the check part and the column with degree 1 is used as the second column of the check part. The value of the element in the last row of the remaining columns in the base matrix except the state column and the column with degree 1 can be 0. In addition, the element "x" in the base matrix represents the part to be designed, and the value of "x" can be 0, 1, or 2. It can be seen that the dimensions of the base matrix are different for different source sparsities, but the structure of the base matrix is ​​similar. When the base matrix is ​​subsequently expanded, the present application adopts a two-step expansion method to obtain the check matrix of JSCC-LDPC. Among them, the first step of expansion corresponds to the first expansion factor, and the first expansion factor is 2, the purpose is to eliminate the element "2" in the base matrix. In addition, after the expansion of the check part of the base matrix, it needs to meet the full rank condition under the modulo-2 operation, and the information part of the base matrix is ​​completely punctured during transmission.

[0207] (1) For the JSCC-LDPC basis matrix with source sparsity p1 = 0.05, the basis matrix dimension is [5, 11], i.e., 5 rows and 11 columns. The basis matrix satisfies the following characteristics:

[0208] a) One of the columns in the check part is a status column. For example, the column is the first column in the check part, specifically [1 11 2 2] T ,The specific selection of this column is a compromise between the ,decoding threshold and trap set / ring elimination;

[0209] b) One of the columns of the check part is a column with degree 1. Exemplarily, the column is the second column of the check part matrix, and specifically can be [0 0 0 0 1] T ; This design combined with the aforementioned state column can achieve a very low decoding threshold, and the complexity of encoding and decoding is greatly reduced. It should be noted that the 1 in this column can be located at any position in the column, and the 1 in this column must be connected to the 2 in the state column (rather than connected to 1), or in other words, in the same row in the basis matrix;

[0210] c) Except for the first two columns of the check part, the last row of the remaining part of the basis matrix is ​​all set to 0;

[0211] Although such a design reduces the degree of design freedom, that is, some sacrifices are made in decoding threshold or performance, the complexity, especially the complexity of the encoding end, is lower.

[0212] According to the above constraint C2, in order to optimize the decoding performance of the constructed base matrix under medium and short codes, it is necessary to constrain the density of the base matrix, that is, to control the maximum row weight and column weight. For p1=0.05, this application provides six constraints, respectively expressed as C3R8, C3R9, C4R8, C4R9, C5R8 and C5R9 (some examples of the first setting condition, i.e. C2). Among them, the constraint C3R8 indicates that the maximum column weight containing x in the base matrix does not exceed 3, and the maximum row weight of the entire base matrix does not exceed 8. Other constraints can be derived from this and will not be explained one by one.

[0213] The following are the three base matrices with the minimum decoding threshold under the above six constraints, where Thr represents the decoding threshold of the base matrix provided in this application at the corresponding code rate. The decoding threshold refers to the minimum Eb / N0 (Eb represents the average energy of a binary bit, and N0 represents the noise power spectral density) required for successful decoding of the JSCC-LDPC corresponding to the base matrix when the code length is infinite, and Gap refers to the distance (expressed in dB) between the decoding threshold and the Eb / N0 corresponding to the channel capacity of the corresponding code rate. The meanings of these parameters are also applicable to other embodiments below and will not be repeated.

[0214] In addition, the three basis matrices with the smallest decoding threshold under these constraints (as some examples of candidate basis matrices) can be used as candidate basis matrices of JSCC-LDPC to expand and obtain the final check matrix.

[0215] C3R8:

[0216]

[0217]

[0218]

[0219] C4R8:

[0220]

[0221]

[0222]

[0223] C5R8:

[0224]

[0225]

[0226]

[0227] C3R9:

[0228]

[0229]

[0230]

[0231] C4R9:

[0232]

[0233]

[0234]

[0235] C5R9:

[0236]

[0237]

[0238]

[0239] The basis matrices given above are actually examples of candidate basis matrices in this application. For a candidate basis matrix, it can be expanded in two steps to obtain a check matrix, and in the expansion process, a basis matrix can be expanded to obtain multiple check matrices (which can be considered as candidate check matrices). Therefore, based on multiple candidate basis matrices, multiple candidate check matrices will be obtained. Next, by performing performance evaluation on the candidate check matrix, a candidate check matrix whose decoding performance meets the requirements (e.g., optimal decoding performance) is selected as the final check matrix. As an example, assuming that the final JSCC-LDPC check matrix a is obtained based on the expansion of the candidate basis matrix b, the candidate basis matrix b is the basis matrix provided in this application, and the check matrix a is the check matrix provided in this application.

[0240] Exemplarily, the base matrix corresponding to p1=0.05 may be the following matrix 1, or a matrix obtained by performing row transformation and / or column transformation on matrix 1:

[0241] Matrix 1

[0242]

[0243] It should be noted that in order to obtain a check matrix with ideal performance, a large number of candidate basis matrices are used to determine the final check matrix. Due to limited space, the above embodiment only provides three basis matrices with the smallest decoding thresholds as candidate basis matrices for the six constraints. Of course, candidate basis matrices can also be provided for other constraints different from the above six constraints, and a larger number of basis matrices can be provided as candidate basis matrices for each constraint. For example, the first 100 basis matrices in the order of the decoding threshold from small to large are provided as candidate basis matrices for each constraint, without limitation.

[0244] Based on a candidate base matrix, the first step is to expand the candidate base matrix with an expansion factor z=2 (i.e., the first expansion factor) to obtain a first matrix. Among them, the element "2" in the candidate base matrix can be expanded to a 2×2 all-1 matrix, and the element "1" can be expanded to a 2×2 unit matrix or a 2×2 anti-diagonal square matrix; the second step is to expand the first matrix with an expansion factor z=32 (i.e., an example of the second expansion factor) to obtain a candidate check matrix. Next, the candidate check matrix can be subjected to trap set analysis to evaluate the decoding performance of the candidate check matrix.

[0245] By evaluating the decoding performance of multiple candidate check matrices, the check matrix of JSCC-LDPC can be finally selected from the multiple candidate check matrices. Exemplarily, the above matrix 1 is expanded to obtain the check matrix corresponding to p1=0.05 provided in this application. The check matrix can be specifically such as matrix 2, or a matrix obtained by row transformation and / or column transformation of matrix 2. Matrix 2 can refer to the above embodiment and will not be listed again.

[0246] (2) For the basis matrix of JSCC-LDPC with source sparsity p1 = 0.11, the dimension of the basis matrix is ​​[6,11], i.e., 6 rows and 11 columns. The main features of the basis matrix are described as follows:

[0247] a) One of the columns in the check part is a status column. For example, this column is the first column in the check part, specifically [1 212 1 2] T ,The specific selection of this column is a compromise between the ,decoding threshold and trap set / ring elimination;

[0248] b) One of the columns of the check part is a column with degree 1. Exemplarily, the column is the second column of the check part matrix, and specifically can be [0 0 0 0 0 1] T ; This design combined with the aforementioned state column can achieve a very low decoding threshold, and the complexity of encoding and decoding is greatly reduced. Similarly, the 1 in this column can be located at any position in the column, and the 1 in this column must be connected to the 2 in the state column, or in the same row in the matrix;

[0249] c) Except for the first two columns of the check part, the last row of the remaining part of the basis matrix can be set to all zeros.

[0250] According to the constraint condition C2, in order to optimize the decoding performance of the constructed base matrix under medium and short codes, it is necessary to further constrain the density of the base matrix, that is, to control the maximum row weight and column weight.

[0251] For p1=0.11, this application provides six constraints, which are respectively expressed as C3R6, C3R7, C4R6, C4R7, C5R7 and C5R8. C3R6 means that the maximum column weight containing "x" in the basis matrix does not exceed 3, and the maximum row weight of the entire basis matrix does not exceed 6. The remaining constraints are similar to C3R6 and will not be described one by one.

[0252] The following are three basis matrices with the smallest decoding thresholds under the above six constraints.

[0253] C3R6:

[0254]

[0255]

[0256]

[0257] C4R6:

[0258]

[0259]

[0260]

[0261] C5R7:

[0262]

[0263]

[0264]

[0265] C3R7:

[0266]

[0267]

[0268]

[0269] C4R7:

[0270]

[0271]

[0272]

[0273] C5R8:

[0274]

[0275]

[0276]

[0277] Exemplarily, the p1=0.11 provided in the present application corresponds to a base matrix such as the following matrix 3, or a matrix obtained by performing row transformation and / or column transformation on matrix 3:

[0278] Matrix 3

[0279]

[0280] Based on the base matrix, the first step is to expand the base matrix with an expansion factor z=2 to obtain a first matrix, wherein the element "2" in the base matrix can be expanded to a 2×2 all-1 matrix, and the element "1" can be expanded to a 2×2 unit matrix or a 2×2 anti-diagonal matrix; the second step is to expand the first matrix with an expansion factor z=32 (i.e., an example of the second expansion factor) to obtain a check matrix.

[0281] The matrix obtained by expanding matrix 3 is the check matrix corresponding to p1=0.11 provided in this application, which can be matrix 4 or a matrix obtained by row transformation and / or column transformation of matrix 4. Matrix 4 can refer to the above embodiment and will not be listed again.

[0282] (3) For the JSCC-LDPC basis matrix with source sparsity p1 = 0.30, the dimension of the basis matrix is ​​[7,11], i.e., 7 rows and 11 columns. The main features of the basis matrix are described as follows:

[0283] a) One of the columns of the check matrix is ​​a state column. Exemplarily, the column is the first column of the check matrix, specifically [1 1 1 1 2 2 2] T ,The specific selection of this column is a compromise between the ,decoding threshold and trap set / ring elimination;

[0284] b) One of the columns of the check matrix is ​​a column with a degree of 1. Exemplarily, the column is the second column of the check matrix, and specifically may be [0 0 0 0 0 0 1] T; This design combined with the aforementioned state column can achieve a very low decoding threshold, and the complexity of encoding and decoding is greatly reduced. The 1 in this column can be located at any position in the column, and the 1 in this column must be connected to the 2 in the state column, or be in the same row in the matrix;

[0285] c) Except for the first two columns of the check matrix, the last row of the remaining part of the basis matrix is ​​all set to 0.

[0286] According to the constraint C2, in order to optimize the decoding performance of the constructed base matrix under medium and short codes, it is necessary to further constrain the density of the base matrix. For p1=0.30, this application provides eight constraints, which are respectively expressed as C3R6, C3R7, C4R6, C4R7, C4R8, C5R7, C5R8 and C5R9. The following are the three base matrices with the smallest decoding threshold under these eight constraints.

[0287] C3R6:

[0288]

[0289]

[0290]

[0291] C4R6:

[0292]

[0293]

[0294]

[0295] C4R8:

[0296]

[0297]

[0298]

[0299] C5R8:

[0300]

[0301]

[0302]

[0303] C3R7:

[0304]

[0305]

[0306]

[0307] C4R7:

[0308]

[0309]

[0310]

[0311] C5R7:

[0312]

[0313]

[0314]

[0315] C5R9:

[0316]

[0317]

[0318]

[0319] Exemplarily, the base matrix corresponding to p1=0.30 given in the present application is the following matrix 5, or a matrix obtained by performing row transformation and / or column transformation on matrix 5.

[0320] Matrix 5

[0321]

[0322] Based on the base matrix (i.e., matrix 5), the first step is to expand the base matrix with an expansion factor z=2 (i.e., the first expansion factor) to obtain a first matrix. Among them, the element "2" in the base matrix can be expanded to a 2×2 all-1 matrix, and the element "1" can be expanded to a 2×2 unit matrix or a 2×2 anti-diagonal square matrix; the second step is to expand the first matrix with an expansion factor z=32 (i.e., an example of the second expansion factor) to obtain a candidate check matrix.

[0323] The matrix obtained based on matrix 5 is the check matrix corresponding to p1=0.30 given in this application, which can be matrix 6, or a matrix obtained by row transformation and / or column transformation of matrix 6. Matrix 6 can refer to the above embodiment and will not be listed again here.

[0324] Optionally, in each embodiment, the positions of the information part and the check part of the base matrix or the check matrix can be interchanged. For example, in the base matrix, the information part is on the left and the check part is on the right; or, the check part is on the left and the information part is on the right, without limitation.

[0325] Combine the following Figure 14 to Figure 16 The decoding performance of JSCC-LDPC provided by this application under source sparsity p1 is given and compared with the performance of SSCC. Among them, SSCC is a traditional coding and decoding scheme with separate source coding and channel coding. Here, the source coding uses classic traditional arithmetic coding for source coding (lossy), and the channel coding uses the NR LDPC code in the 5G standard, and the maximum number of iterations is set to 20 times; in JSCC, p1 is the source sparsity, which represents the probability of "1" appearing in each source bit.

[0326] Figures 14 to 16 The block error rate (BLER) and signal noise ratio (SNR) curves are shown in FIG. 1 . It can be seen that, compared with the traditional SSCC scheme, the BLER of the technical scheme of the present application is the lowest under the same SNR; in other words, under the same BLER, the technical scheme of the present application has the lowest requirement for SNR.

[0327] according to Figure 14-16 It can be seen from the simulation results that the JSCC-LDPC provided in this application can achieve error control performance better than the traditional SSCC scheme under various typical p1s, so the decoding performance is better than the traditional SSCC scheme.

[0328] Specifically, in Fig.14 In the example, “Ra=1.5, JSCC-LDPC” indicates the JSCC scheme of the present application: a binary biased signal source N with a length of 384 bits S , encoded as a codeword N with a length of 256 bits C , the code rate is 1.5, and the matrix 2 given in this application is used as the check matrix. "Ra=1.5, SSCC (NR LDPC, Rcc=0.78)" represents the SSCC scheme, and the length of the binary biased signal source N is 384 bits. S It is first compressed to 200 bits by arithmetic coding (arithmetic coding is a source coding, source coding, SC), and then NR LDPC coding is used to obtain 256 bits of N C , the code rate of the channel coding part is 0.78, and the overall code rate is the same as the JSCC scheme, which is 1.5. "Ra=1.5,SSCC(NR LDPC,Rcc=0.81)" indicates the SSCC scheme, with a length of 384 bits of binary biased signal source NS It is first compressed to 208 bits through arithmetic coding, and then NR LDPC coding is performed to obtain 256 bits of N C , the code rate of the channel coding part is 0.81, and the overall code rate is the same as the JSCC solution, which is 1.5. Obviously, it can be seen that compared with the traditional SSCC solution, the BLER of the technical solution of the present application is the lowest under the same SNR; or in other words, under the same BLER, the technical solution of the present application has the lowest requirement for SNR.

[0329] exist Fig.15 In the figure, "Ra=1.0, JSCC-LDPC" represents the JSCC scheme of the present application, where Ra=1.0 means the code rate is 1.0, and the matrix 4 provided in the present application is used as the check matrix. Obviously, it can be seen that compared with the traditional SSCC scheme, the BLER of the present application's technical scheme is the lowest under the same SNR; in other words, under the same BLER, the technical scheme of the present application has the lowest requirement for SNR.

[0330] exist Fig.16 In the figure, "Ra=0.67, JSCC-LDPC" indicates the JSCC scheme of the present application, where Ra=0.67 indicates that the code rate is 0.67 (i.e., 2 / 3), and the matrix 6 provided in the present application is used as the check matrix. Obviously, it can be seen that compared with the traditional SSCC scheme, the BLER of the present application's technical scheme is the lowest under the same SNR; in other words, under the same BLER, the technical scheme of the present application has the lowest requirement for SNR.

[0331] Based on Fig.14 The simulation parameters and simulation results of the present invention can be understood by those skilled in the art. Fig.15 and Fig.16 The simulation data and simulation results are similar and will not be repeated.

[0332] The encoding method provided in the embodiment of the present application is described in detail above. The communication device provided in the present application is introduced below.

[0333] See also Fig.17 , the present application provides a communication device 1000.

[0334] The communication device 1000 may be a coding device, or may be a communication device applied to a coding device capable of implementing the method executed by the above-mentioned coding device, such as a chip, a chip system or a circuit.

[0335] Alternatively, the communication device 1000 may be a decoding device, or a communication device applied to a decoding device and capable of implementing the method executed by the above-mentioned decoding device, such as a chip, a chip system or a circuit.

[0336] like Fig.17 The communication device 1000 includes a processing module 1001, which may also be referred to as a processor, a processing board, a processing unit, or a processing device.

[0337] When the communication device 1000 is a coding device, or a communication device applied to a coding device and capable of implementing the method performed by the coding device, the processing module 1001 is used to encode based on a check matrix, and the details of the check matrix refer to the description of the check matrix in the above method, which will not be repeated here. When the communication device 1000 is a decoding device, or a communication device applied to a decoding device and capable of implementing the method performed by the decoding device, the processing module 1001 is used to decode based on a check matrix, and the details of the check matrix refer to the description of the check matrix in the above method, which will not be repeated here.

[0338] Optionally, the communication device 1000 further includes a communication module 1002, which may also be referred to as a transceiver module, a transceiver, a transceiver, or a transceiver device. When the communication device 1000 is a coding device, or a communication device applied to a coding device and capable of implementing the method performed by the coding device, the communication module 1002 may be used to implement Figure 5 to Figure 13 The transceiver function of the encoding device in each embodiment; when the communication device 1000 is a decoding device, or a communication device applied to a decoding device and capable of implementing the method executed by the above decoding device, the communication module 1002 can be used to implement Figure 5 to Figure 13 The transceiver function of the decoding device in each of the above embodiments.

[0339] In addition, it should be noted that the aforementioned communication module and / or processing module can be implemented by a virtual module, for example, the processing module can be implemented by a software function unit or a virtual device, and the communication module can be implemented by a software function or a virtual device. Alternatively, the processing module or the communication module can also be implemented by a physical device, for example, if the device is implemented using a chip / circuit (such as an integrated circuit or a logic circuit, etc.). The communication module can be an input-output circuit and / or a communication interface, performing input operations (corresponding to the aforementioned receiving operations) and output operations (corresponding to the aforementioned sending operations); the processing module is an integrated processor or microprocessor or circuit (such as an integrated circuit, a logic circuit, etc.).

[0340] The division of modules in this application is schematic and is only a logical function division. There may be other division methods in actual implementation. In addition, each functional module in each example of this application may be integrated into one processor, or may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated modules may be implemented in the form of hardware or in the form of software functional modules.

[0341] In addition, see Fig.18 The present application also provides a communication device 1100. Optionally, the communication device 1100 may be a chip or a chip system. Optionally, in the present application, the chip system may be composed of a chip, or may include a chip and other discrete devices.

[0342] The communication device 1100 includes at least one processor 1110, which implements the functions of the encoding device or the decoding device described in the above examples.

[0343] Optionally, the processor 1110 is coupled to a memory, and the memory may be located within the communication device, or the memory may be integrated with the processor, or the memory may be located outside the communication device. For example, the communication device 1100 may also include at least one memory 1120. The memory 1120 stores the computer programs, instructions, or data necessary for implementing any of the above method embodiments; the processor 1110 may execute the computer programs, instructions, or data stored in the memory 1120 to complete the method of any of the above embodiments.

[0344] Optionally, the communication device 1100 may further include a communication interface 1130, and the communication device 1100 may exchange information with other devices through the communication interface 1130. Exemplarily, the communication interface 1130 may be a transceiver, a circuit, a bus, a module, a pin, or other types of communication interfaces. When the communication device 1100 is a chip-type device or circuit, the communication interface 1130 in the device 1100 may also be an input-output circuit, which may input information (or receive information) and output information (or send information), and the processor 1110 may be an integrated processor, a microprocessor, an integrated circuit or a logic circuit, etc., and the processor may determine the output information based on the input information.

[0345] The coupling in this application is an indirect coupling or communication connection between devices, units or modules, which can be electrical, mechanical or other forms, and is used for information exchange between devices, units or modules. The processor 1110 may cooperate with the memory 1120 and the communication interface 1130. The specific connection medium between the above-mentioned processor 1110, memory 1120 and communication interface 1130 is not limited in this application.

[0346] Alternatively, if Fig.18 As shown in , the processor 1110, the memory 1120 and the communication interface 1130 are interconnected via a bus 1140. Optionally, the bus may include an address bus, a data bus, a control bus and other types of buses. In addition, for ease of representation, Fig.18 One bus 1140 is shown, but this does not mean that there is only one bus or one type of bus.

[0347] In this application, the processor may be a general-purpose processor, a digital signal processor, an application-specific integrated circuit, a field programmable gate array or other programmable logic device, a discrete gate or transistor logic device, or a discrete hardware component, and may implement or execute the methods, steps, and logic block diagrams disclosed in this application. A general-purpose processor may be a microprocessor or any conventional processor, etc. The steps of the method disclosed in this application may be directly embodied as being executed by a hardware processor, or may be executed by a combination of hardware and software modules in the processor.

[0348] The memory may be a non-volatile memory, such as a hard disk drive (HDD) or a solid-state drive (SSD), or a volatile memory, such as a random-access memory (RAM). The memory is any other medium that can be used to carry or store the desired program code in the form of instructions or data structures and can be accessed by a computer, but is not limited thereto. The memory in the present application may also be a circuit or any other device that can realize a storage function, for storing program instructions and / or data.

[0349] Since the communication device 1100 provided in this example can be applied to an encoding device to complete a method executed by the encoding device, or the communication device 1100 can be applied to a decoding device to complete a method executed by the decoding device, the technical effects that can be obtained by the device embodiment can refer to the description in the above method embodiment, and will not be repeated here.

[0350] Based on the above examples, the present application provides a communication system, including the encoding device and the decoding device in the above method embodiment.

[0351] The technical solution provided in this application can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in this application is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, a terminal device, an access network device or other programmable device. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another computer-readable storage medium. For example, the computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital video disc (DVD)), or a semiconductor medium, etc.

[0352] In the present application, under the premise of no logical contradiction, the examples may reference each other, for example, the methods and / or terms between method embodiments may reference each other, for example, the functions and / or terms between device embodiments may reference each other, for example, the functions and / or terms between device examples and method examples may reference each other.

[0353] The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

[0354] In addition, each functional unit in each embodiment of the present application may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

[0355] The multiple (items) involved in the embodiments of the present application refer to two (items) or more than two (items). "And / or" describes the association relationship of associated objects, indicating that three relationships may exist. For example, A and / or B can represent: A exists alone, A and B exist at the same time, and B exists alone. In addition, it should be understood that although the terms first, second, etc. may be used to describe each object in this application, these objects should not be limited to these terms. These terms are only used to distinguish each object from each other.

[0356] The term "comprising" and any variation thereof mentioned in the embodiments of the present application are intended to cover non-exclusive inclusions. For example, a process, method, system, product or equipment comprising a series of steps or units is not limited to the listed steps or units, but optionally also includes other steps or units that are not listed, or optionally also includes other steps or units inherent to these processes, methods, products or equipment. In addition, words such as "exemplarily" or "for example" are used to represent examples, illustrations or explanations. Any method or design described as "exemplarily" or "for example" in the present application should not be interpreted as being more preferred or more advantageous than other methods or designs. Specifically, the use of words such as "exemplarily" or "for example" is intended to illustrate related technical solutions or concepts in a concrete manner.

[0357] Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments and will not be repeated here.

[0358] In the several embodiments provided in the present application, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

[0359] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present application can be essentially or partly embodied in the form of a software product that contributes to the prior art. The computer software product is stored in a storage medium and includes several instructions for a computer device (which can be a personal computer, a server, or a network device, etc.) to perform all or part of the steps of the methods described in the various embodiments of the present application. The aforementioned storage media include: U disk, mobile hard disk, read-only memory (ROM), random access memory (RAM), disk or optical disk, and other media that can store program codes.

[0360] The above is only a specific implementation of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art who is familiar with the present technical field can easily think of changes or substitutions within the technical scope disclosed in the present application, which should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A method for encoding or decoding in a wireless system, characterized in that: include: Encoding or decoding based on a check matrix; The check matrix is ​​obtained by expanding a base matrix, the base matrix includes an information part and a check part, the check part includes a state column and a column with a degree of 1, the state column includes at least one element "2", the column with a degree of 1 includes an element "1", and an element "2" in the at least one element "2" and the element "1" are located in the same row of the base matrix.

2. According to the method of claim 1, the status column also includes an element "1", and the number of elements "2" and element "1" in the status column is close to that of the elements "1" in the status column.

3. The method according to claim 1 or 2, characterized in that Elements of the remaining columns of the base matrix except the state column and the column of degree 1 include element "2", element "1" or element "0".

4. The method according to any one of claims 1 to 3, characterized in that The last row of the remaining columns of the base matrix except the state column and the column with degree 1 is an element "0".

5. The method according to any one of claims 1 to 4, characterized in that The elements of the state column are variable nodes corresponding to the punctured check bits.

6. The method according to any one of claims 1 to 5, characterized in that The expansion includes two-step expansion, the first expansion factor corresponding to the first expansion of the two-step expansion is 2, and the first expansion includes: The base matrix is ​​expanded based on the first expansion factor to obtain a first matrix, wherein the element "2" in the base matrix is ​​expanded to a 2×2 all-1 matrix, the element "1" is expanded to a 2×2 unit matrix or an anti-diagonal square matrix, and the element "0" is expanded to a 2×2 all-zero square matrix.

7. The method according to claim 6, characterized in that The second expansion factor corresponding to the second expansion step of the two-step expansion is 32, and the second expansion step includes: The first matrix is ​​expanded based on the second expansion factor to obtain the check matrix, and the element "1" in the first matrix is ​​expanded to a 32×32 circulant permutation matrix P i , the element "0" in the first matrix is ​​expanded to a 32×32 all-zero square matrix, where 0≤i≤31 and i is an integer, P0 represents the unit matrix, P i It represents the square matrix obtained by cyclically shifting right each column vector in P0 by i positions.

8. The method according to any one of claims 1 to 7, characterized in that The base matrix includes the following matrix 1, or a matrix obtained by performing row transformation or column transformation on matrix 1; Matrix 1 9. The method according to any one of claims 1 to 8, characterized in that The check matrix includes the following matrix 2, or a matrix obtained by performing row transformation or column transformation on the matrix 2: Matrix 2 The element "-1" in the matrix 2 represents a z×z all-zero matrix, the element "0" represents a z×z unit matrix, and the element P i represents a z×z cyclic permutation matrix, i represents a cyclic shift value, 0≤i≤z-1, i is an integer, and z represents the second expansion factor.

10. The method according to claim 8 or 9, characterized in that: The sparsity of the basis matrix corresponding to the source belongs to the sparsity interval (0, 0.05].

11. The method according to any one of claims 1 to 7, characterized in that The base matrix includes the following matrix 3, or a matrix obtained by performing row transformation or column transformation on the matrix 3: Matrix 3 12. The method according to any one of claims 1 to 7, 11, characterized in that: The check matrix includes the following matrix 4, or a matrix obtained by performing row transformation or column transformation on the matrix 4: Matrix 4 The element "-1" in the matrix 4 represents a z×z all-zero matrix, the element "0" represents a z×z unit matrix, and the element P i represents a z×z cyclic permutation matrix, i represents a cyclic shift value, 0≤i≤z-1, i is an integer, and z represents the second expansion factor.

13. The method according to claim 11 or 12, characterized in that: The sparsity of the basis matrix corresponding to the source belongs to the sparsity interval (0.05, 0.11].

14. The method according to any one of claims 1 to 7, characterized in that The base matrix includes the following matrix 5, or a matrix obtained by performing row transformation or column transformation on the matrix 5: Matrix 5 15. The method according to any one of claims 1 to 7, 14, characterized in that: The check matrix includes the following matrix 6, or a matrix obtained by performing row transformation or column transformation on the matrix 6: Matrix 6 The element "-1" in the matrix 6 represents a z×z all-zero matrix, the element "0" represents a z×z unit matrix, and the element P i represents a z×z cyclic permutation matrix, i represents a cyclic shift value, 0≤i≤z-1, i is an integer, and z represents the second expansion factor.

16. The method according to claim 14 or 15, characterized in that The sparsity of the basis matrix corresponding to the source belongs to the sparsity interval (0.11, 0.3].

17. A communication device in a wireless system, characterized in that: include: A processing module, used for encoding based on a check matrix; The check matrix is ​​obtained by expanding a base matrix, the base matrix includes an information part and a check part, the check part includes a state column and a column with a degree of 1, the state column includes at least one element "2", the column with a degree of 1 includes an element "1", and an element "2" in the at least one element "2" and the element "1" are located in the same row of the base matrix.

18. The communication device according to claim 17, wherein the status column further comprises an element "1", and the number of elements "2" and elements "1" in the status column is close to that of the elements "1" in the status column.

19. The communication device according to claim 17 or 18, characterized in that: Elements of the remaining columns of the base matrix except the state column and the column of degree 1 include element "2", element "1" or element "0".

20. The communication device according to any one of claims 17 to 19, characterized in that: The last row of the remaining columns of the base matrix except the state column and the column with degree 1 is an element "0".

21. The communication device according to any one of claims 17 to 20, characterized in that: The elements of the state column are variable nodes corresponding to the punctured check bits.

22. The communication device according to any one of claims 17 to 21, characterized in that: The expansion includes two-step expansion, the first expansion factor corresponding to the first expansion of the two-step expansion is 2, and the first expansion includes: The base matrix is ​​expanded based on the first expansion factor to obtain a first matrix, wherein the element "2" in the base matrix is ​​expanded to a 2×2 all-1 matrix, the element "1" is expanded to a 2×2 unit matrix or an anti-diagonal square matrix, and the element "0" is expanded to a 2×2 all-zero square matrix.

23. The communication device according to claim 22, characterized in that: The second expansion factor corresponding to the second step expansion is 32, and the second step expansion includes: The first matrix is ​​expanded based on the second expansion factor to obtain the check matrix, and the element "1" in the first matrix is ​​expanded to a 32×32 circulant permutation matrix P i , the element "0" in the first matrix is ​​expanded to a 32×32 all-zero square matrix, where 0≤i≤31 and i is an integer, P0 represents the unit matrix, P i It represents the square matrix obtained by cyclically shifting right each column vector in P0 by i positions.

24. A communication device in a wireless system, characterized in that: include: A processing module, used for decoding based on a check matrix; The check matrix is ​​obtained by expanding a base matrix, the base matrix includes an information part and a check part, the check part includes a state column and a column with a degree of 1, the state column includes at least one element "2", the column with a degree of 1 includes an element "1", and an element "2" in the at least one element "2" and the element "1" are located in the same row of the base matrix.

25. The communication device according to claim 24, wherein the status column further comprises an element "1", and the number of elements "2" and elements "1" in the status column is close to that of the elements "1" in the status column.

26. The communication device according to claim 24 or 25, characterized in that: Elements of the remaining columns of the base matrix except the state column and the column of degree 1 include element "2", element "1" or element "0".

27. The communication device according to any one of claims 24 to 26, characterized in that: The last row of the remaining columns of the base matrix except the state column and the column with degree 1 is an element "0".

28. The communication device according to any one of claims 24 to 27, characterized in that: The elements of the state column are variable nodes corresponding to the punctured check bits.

29. The communication device according to any one of claims 24 to 28, characterized in that: The expansion includes two-step expansion, the first expansion factor corresponding to the first expansion of the two-step expansion is 2, and the first expansion includes: The base matrix is ​​expanded based on the first expansion factor to obtain a first matrix, wherein the element "2" in the base matrix is ​​expanded to a 2×2 all-1 matrix, the element "1" is expanded to a 2×2 unit matrix or an anti-diagonal square matrix, and the element "0" is expanded to a 2×2 all-zero square matrix.

30. The communication device according to claim 29, characterized in that The second expansion factor corresponding to the second step expansion is 32, and the second step expansion includes: The first matrix is ​​expanded based on the second expansion factor to obtain the check matrix, and the element "1" in the first matrix is ​​expanded to a 32×32 circulant permutation matrix P i , the element "0" in the first matrix is ​​expanded to a 32×32 all-zero square matrix, where 0≤i≤31 and i is an integer, P0 represents the unit matrix, P i It represents the square matrix obtained by cyclically shifting right each column vector in P0 by i positions.

31. A communication device, characterized in that: It comprises a communication interface and a circuit, wherein the communication interface is used to decode an information sequence to be encoded and input the information sequence to be encoded into the circuit; the circuit is used to execute the method as described in any one of claims 1 to 16, encode the information sequence to be encoded based on a check matrix to obtain coded bits; the communication interface is also used to output the coded bits.

32. A communication device, characterized in that: It comprises a communication interface and a circuit, wherein the communication interface is used to decode a sequence to be decoded and input the sequence to be decoded into the circuit; the circuit is used to execute the method according to any one of claims 1 to 16, decode the bits to be decoded based on a check matrix to obtain a decoded sequence; the communication interface is also used to output the decoded sequence.

33. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores computer instructions. When the computer instructions are executed on a computer, the method according to any one of claims 1 to 16 is implemented.

34. A computer program product, characterized in that The computer program product comprises a computer program code, and when the computer program code is run on a computer, the method according to any one of claims 1 to 16 is implemented.

Images